file-organizer
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill generates shell commands such as
find,mv,mkdir, andduto analyze and modify the local filesystem. These operations are performed on user-specified paths, which could lead to command injection or directory traversal if pathnames are not handled with strict shell escaping by the agent. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection from the filesystem. Ingestion points: The agent reads directory structures and filenames via
lsandfind. Boundary markers: No explicit delimiters are defined to separate file metadata from agent instructions. Capability inventory: The agent can move files, rename files, and create directories. Sanitization: No sanitization of filenames or paths is mentioned before processing. - [DATA_EXFILTRATION]: The skill performs deep inspection of files, including calculating MD5 hashes and identifying file types via the
filecommand. While no network calls are present, this exposes a high level of metadata about the user's private data to the agent context.
Audit Metadata